Product inhibition in nucleophilic aromatic substitution through DPPPent-supported π-arene catalysis

Article abstract of DOI:10.1039/d0dt00786b

Nucleophilic aromatic substitution (SNAr) of fluorobenzene by morpholine at a bis(diphenylphosphino)pentane-supported ruthenim complex is investigated as a model system for π-arene catalysis through the synthesis and full characterization of proposed intermediates. The SNAr step proceeds quickly at room temperature, however the product N-phenylmorpholine binds tightly to the ruthenium ion. In the case examined, the thermodynamics of arene binding favor product N-phenylmorpholine over fluorobenzene binding by a factor of 2000, corresponding to significant product inhibition. Observations of the catalyst resting state support this hypothesis and demonstrate an additive-controlled role for a previously-proposed ligand cyclometalation. This journal is

Full text of DOI:10.1039/d0dt00786b

Dalton
Transactions
View Article Online
View Journal
PAPER
Product inhibition in nucleophilic aromatic
substitution through DPPPent-supported
Cite this: DOI: 10.1039/d0dt00786b
π-arene catalysis†
Benjamin R. J. Mueller
and Nathan D. Schley
*
Nucleophilic aromatic substitution (SNAr) of fluorobenzene by morpholine at a bis(diphenylphosphino)
pentane-supported ruthenim complex is investigated as a model system for π-arene catalysis through the
synthesis and full characterization of proposed intermediates. The SNAr step proceeds quickly at room
temperature, however the product N-phenylmorpholine binds tightly to the ruthenium ion. In the case
examined, the thermodynamics of arene binding favor product N-phenylmorpholine over fluorobenzene
binding by a factor of 2000, corresponding to significant product inhibition. Observations of the catalyst
resting state support this hypothesis and demonstrate an additive-controlled role for a previously-pro-
posed ligand cyclometalation.
Received 2nd March 2020,
Accepted 6th July 2020
DOI: 10.1039/d0dt00786b
rsc.li/dalton
180 °C. Among these, the phosphine-supported ruthenium(II)
catalysts have been more successful for the SNAr of fluoroar-
Introduction
Nucleophilic aromatic substitution (SNAr) of haloarenes is a enes by amines.13–15 In all cases, the reaction is speculated to
powerful synthetic tool that finds wide use in organic chem- follow a general mechanism proposed by Semmelhack et al.17
istry. A major limitation of this reaction, however, is the (Scheme 1) wherein arene exchange allows for catalytic turn-
requirement for electron deficient arene electrophiles. One over after SNAr. Electron-deficient arenes have poorer arene
strategy for the activation of otherwise unactivated arenes is binding thermodynamics while electron-rich arenes have poor
through η6 binding to a metal center, which gives π arene com- arene exchange kinetics.18 This ensures that product inhi-
plexes with significantly enhanced electrophilicity.1–6 SNAr bition is an intrinsic challenge in all cases where the product
reactions of η6 haloarenes often proceed at room temperature arene is more electron rich than the haloarene starting
even for substrates like chlorobenzene7–9 which is largely inert material, though to our knowledge this has never been quanti-
absent metal-ion activation. In principle, catalytic turnover can fied in a catalytic system.
be achieved by product arene exchange for the starting
The potential complementarity of catalytic SNAr to better-
material haloarene as depicted in Scheme 1. However in most developed cross-coupling methods has encouraged our
cases the strong binding of the arene to the metal requires research group to examine this class of transformations in
photolytic or oxidative conditions to liberate the product,10 more detail. We chose to begin with a mechanistic study of a
which has largely precluded catalytic applications with rare 1,5-bis(diphenylphosphino)pentane (DPPPent)-supported Ru
exceptions.
catalyst for SNAr reported by the Shibata group (eqn (1)).15 This
All existing examples of catalytic SNAr reactions involving catalytic system is closely related to one applied in the catalytic
η6-arene coordination are limited to 2nd row transition metal anti-Markovnikov hydroamination of styrene by the Hartwig
catalysts. A rhodium(III) example11 and a limited number of
ruthenium(II) complexes – two containing cyclopentadienyl
derivatives and three containing phosphine ligands, have been
shown to serve as catalysts for SNAr of haloarenes by fluoride12
and amines13–16 at temperatures ranging from 100 °C to
Department of Chemistry, Vanderbilt University, Nashville, Tennessee 37235, USA.
E-mail: nathan.schley@vanderbilt.edu
†Electronic supplementary information (ESI) available. CCDC 1960834, 1960835,
1960836 and 1987630. For ESI and crystallographic data in CIF or other elec-
tronic format see DOI: 10.1039/d0dt00786b
Scheme 1 Reported catalysts for π-arene SNAr.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
group.19 In that study Hartwig was able to show that DPPPent Catalyst resting state and the role of additives
undergoes cyclometalation to give a facial, tridentate ligand-
Under the previously reported catalytic conditions, both 1 and
supported ruthenium complex that binds η6 arenes.19
2 catalyze the reaction between fluorobenzene and morpholine
in similar yields to the catalyst generated in situ from Ru(cod)
(methallyl)2 (Table 1). The optimized conditions reported by
Shibata include triethylamine and triethylsilane additives in
stoichiometric amounts which they show are required for high
yields. This observation is born out in our own studies; in the
absence of additives the product is still formed but in reduced
yield (46% vs. 93% with additives). To address the possibility
that reaction additives might influence catalyst speciation, the
identity of the catalyst resting state was investigated by 31P{1H}
NMR both in the presence and absence of silane and amine
additives.
Based on the proposed mechanism put forth by
Semmelhack17 and our own arene binding measurements (vide
infra), the bound phenylmorpholine compound (2) would be
the expected resting state. Indeed, 2 is observed as the catalyst
resting state by 31P{1H} NMR in the absence of triethylsilane
and triethylamine, confirming that it represents a relevant
system for mechanistic experiments (vide infra). However, in
the presence of these additives (the reported optimized catalytic
conditions), 2 is observed only at very short reaction times.
Instead a second, previously unknown species 3 is observed as
the major species during productive catalysis. Initial attempts
to characterize 3 revealed that triethylsilane is necessary for its
formation and that 3 possesses a metal hydride which res-
onates upfield at −9.5 ppm. Analysis of a single-crystal of 3
obtained by careful isolation from a variation of a catalytic
reaction (eqn (4)) revealed that 3 is a bis(phosphine)ruthenium
hydride lacking the alkyl ligand resulting from backbone cyclo-
metalation in 2 (Fig. 2). In separate experiments we found that
3 can be formed by treatment of 2 with 20 equiv. of triethyl-
silane, suggesting a route for the conversion of κ3 cyclometa-
lated complexes to the κ2 form observed in 3.
When applied to SNAr catalysis, DPPPent gave a complex
(generated in situ) that displayed the highest turnover
numbers and mildest reaction conditions of any inter-
molecular π-arene SNAr reaction at the time.15 Their prelimi-
nary mass spectrometry and 31P{1H} NMR experiments suggest
that ruthenium arene complexes analogous to those character-
ized by Hartwig may be formed in situ, which represented an
ideal starting point for further study.
Results and discussion
On the basis of mass spectrometric data, Shibata proposed
π-arene intermediates15 supported by a cyclometalated κ3
DPPPent ligand analogous to the one observed by Hartwig.19
We undertook the synthesis of two arene derivatives bearing a
κ3 DPPPent ligand in an effort to study their properties in
π-arene SNAr catalysis. Complexes 1 and 2 were synthesized in
a single step from Ru(cod)(methallyl)2 using variations on a
reported procedure (eqn (2) and (3)).19 Both complexes were
characterized by 1H and 31P{1H} NMR and combustion ana-
lysis, and their structures were confirmed by single crystal
X-ray diffraction (Fig. 1).
When complex 3 is used as a precatalyst under Shibata’s
optimized conditions, the N-phenylmorpholine product is
Table 1 Effect of precatalyst and additives on reaction yield
Entry
Catalyst
Additivesa
% yieldb
1
2
3
4
5
6
7
8
In situc
1
2
Et3N, Et3SiH
Et3N, Et3SiH
Et3N, Et3SiH
Et3N, Et3SiH
None
None
None
93
98
92d
>99d
46
3
In situc
1
2
3
57
47d
>99d
None
a 1 equiv. of each additive. b Yield by GC-FID. c 5 mol% Ru(cod)(methal-
lyl)2, 7 mol% DPPPent, 10 mol% TfOH. d 2 and 3 contribute 5% to
total yield, see ESI.†
Fig. 1 ORTEP diagrams of complexes 1 (left) and 2 (right). Ellipsoids are
shown at 50% probability.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
that the change in catalyst resting state from 2 to 3 is
accompanied by increased catalyst solubility.
Having determined that the presence of triethylsilane and
triethylamine additives results in a switch in catalyst resting
state from 2 to 3, we attempted to investigate the corres-
ponding fluorobenzene adduct. Unfortunately, efforts to
prepare a fluoroarene complex analogous to 3 by treatment of
1 with triethylsilane gave complex mixtures of products
without evidence for fluoroarene binding by NMR
spectroscopy.
SNAr kinetics and arene binding thermodynamics
Despite our inability to prepare the fluoroarene partner to
complex 3, the observation that 2 serves as the catalyst resting
state in the absence of additives and leads to a productive cata-
lytic reaction led us to pursue mechanistic studies on the 1/2
pair. In particular, the isolation of complex 1 affords us a
unique opportunity to directly measure the rate of SNAr on a
π-arene in a system with catalytic relevance. Under pseudo-first
order conditions, 1 reacts rapidly with morpholine to give 2
within 10 minutes at 23 °C, corresponding to a kobs of
3.8 × 10−3 s−1 (Fig. 3). The reactivity of complex 1 with morpho-
line at room temperature stands in contrast to the metal-free
reaction of morpholine with even very highly-activated nitro-
fluorobenzenes. 2-Nitrofluorobenzene has been reported to
undergo amination by morpholine at 40 °C,20 while 3-nitro-
fluorobenzene requires heating to 100 °C for 60 hours.21
The high rate of conversion of 1 to 2 observed at 23 °C
suggests that this step is unlikely to be the primary determi-
nant of the overall reaction rate under the reported catalytic
conditions (5 mol% Ru, 24 h, 100 °C). Thus we next examined
the arene exchange step in Scheme 1. Efficient displacement
of product from the metal center is believed to be the most
challenging aspect in the development of catalytic SNAr reac-
tions of π-arenes. Hartwig has previously examined the rate of
displacement of N-phenethylmorpholine by styrene on the
same ruthenium system.19
Fig. 2 ORTEP diagram of complex 3. Ellipsoids are shown at 50%
probability.
obtained in quantitative yield (Table 1, entry 4). This obser-
vation argues that ligand cyclometalation observed in 1 and 2
is not necessary for reactivity. Unlike in the case of complex 2,
the performance of complex 3 does not suffer in the absence
of Et3SiH and Et3N additives (Table 1, entry 8). The Shibata
group has previously hypothesized that the inclusion of tri-
ethylsilane and triethylamine is necessary to sequester hydro-
fluoric acid generated as a byproduct of fluoroarene SNAr. The
observation that silane is not required when 3 is used as a pre-
catalyst argues against this hypothesis for the primary function
of silane in the productive catalytic reaction. Instead its most-
significant function appears to be the switch in ligand binding
mode and thus the catalyst resting state from 2 to 3.
Catalytic reactions conducted without additives (Table 1
entries 5–7, Table 2 entries 4–6) tended to give lower yields
and were observed to deposit a yellow precipitate within the
first several hours of the reaction except when 3 was used as a
catalyst. Filtration and analysis of this precipitate after reaction
completion showed that complex 2 precipitates in 85% yield
with respect to the ruthenium precursor. Precipitation was not
observed in the presence of additives, a result which argues
When complex 1 is treated with free N-phenylmorpholine (2
equiv.) in neat fluorobenzene at 23 °C, no arene exchange is
Table 2 Effect of added product on reaction yield after 2 hours
Entry
Additivesa
Time
N-Phenylmorpholine
% yieldb
1
2
3
4
5
6
Et3N, Et3SiH
Et3N, Et3SiH
Et3N, Et3SiH
None
None
None
2 h
2 h
2 h
2 h
2 h
2 h
0 equiv.
32
10
2
15
1
0.5 equiv.
1.0 equiv.
0 equiv.
0.5 equiv.
1.0 equiv.
Fig. 3 Stoichiometric SNAr reaction of complex 1 with morpholine at
23 °C under pseudo-first order conditions. Inset: Ln[1] vs. time.
Conditions: 0.0127 M 1, 0.127 M morpholine in 4 : 1 dioxane/DMF. See
the ESI† for additional details.
2
a 1 equiv. Et3N, 1 equiv. Et3SiH. b Yield by GC-FID.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 3 Rate of product arene exchange in 2
Entry
Temperature
Rate constant 2 → 1 (kobs)
1
2
3
4
5
65 °C
70 °C
75 °C
80 °C
85 °C
5.6 × 10−5 s−1
1.3 × 10−4 s−1
2.4 × 10−4 s−1
4.8 × 10−4 s−1
1.1 × 10−3 s−1
Fig. 4 Predicted fraction of fluoroarene complex 1 as a function of
turnover number; estimated from Keq
.
2000 times greater affinity than fluorobenzene, leading to
strong product inhibition and (2) that the requirement for elev-
ated reaction temperatures is likely dictated largely by the kine-
tics of arene exchange and the requirement for SNAr on the
minute fraction of catalyst present as 1.27 While comparable
studies have not been performed on related catalysts, all cata-
lytic π-arene alkoxylation and amination systems appear to
achieve no more than ca. 20 TON under reported
conditions.11,13,14
Role of phosphine ligands. Further evidence for the sugges-
tion that cyclometalation is not necessary for the reactivity of
the DPPPent system can be obtained through the substitution
of other phosphine ligands. A number of bidentate phos-
phines are found to give modest catalytic activity (Table 4). For
instance while 2,2′-bis(diphenylphosphino)diphenyl ether
(DPEPhos) can coordinate through the biarylether moiety, it
cannot cyclometalate to give an anionic alkyl donor, but still
gives comparable yields to DPPPent (Table 4).28 Other phos-
phines give reduced but still appreciable yields. The results of
our phosphine comparison when taken together with evidence
showing a silane-controlled resting state of the catalytic reac-
tion suggest that ligand cyclometalation is not a defining
feature of SNAr catalysis by the DPPPent system.
observed. On heating to 100 °C, a stable equilibrium between
1 and 2 is obtained that allows for the determination of an
equilibrium constant Keq = 2 × 103 at 100 °C. Using this experi-
mental equilibrium constant we can predict the ratio of com-
plexes 2 and 1 during catalysis. After a single turnover, the
ratio of 2 to 1 is predicted to be 4 : 1, a value that rises rapidly
to >200 : 1 after 10 turnovers (50% conversion). The predicted
fraction of complex 1 as a function of turnover number is
shown in Fig. 4, and demonstrates the dramatic influence of
strong product binding on the predicted catalyst resting state.
Thus, even under idealized conditions, the proportion of cata-
lyst in the fluoroarene form is predicted to fall by two orders of
magnitude by the time the reaction yield has reached 25%.
Indeed, product added at the beginning of the reaction has
a strong inhibitory effect on catalytic turnover both in the pres-
ence and absence of additives (3 and 2 as resting state respect-
ively). The addition of 0.5 equiv. of N-phenylmorpholine leads
to poor catalyst performance over 2 hours, while addition of a
full equivalent of product inhibits catalysis even more dramati-
cally (Table 2). The additive-free case appears to be affected to
a larger extent, which may stem from the low apparent solubi-
lity of 2 (vide supra).
While the thermodynamics of product binding can be
expected to decrease the fraction in the active form at equili-
brium, the rate of arene exchange should determine whether
equilibrium concentrations are achieved under catalytic con-
ditions. To that end, we examined the rate of displacement of
N-phenylmorpholine by a large excess of fluorobenzene (con-
version of 2 to 1). Initial rate constants for the conversion of 2
to 1 via arene exchange are shown as a function of temperature
in Table 3. Under these conditions, product displacement at
65 °C is found to be two orders of magnitude slower than SNAr
measured at 23 °C. From these data the activation energy of
arene exchange is calculated to be 34 kcal mol−1. The precise
mechanism of arene exchange can be complex and conditions-
dependent,18,22–26 but these values provide some insight into
the lability of the product arene in 2.
Table 4 Effect of phosphine ligand on reaction yield
Phosphine
PPh3
Yielda
Phosphine
Yielda
49%b
93%
PAr3
19%b
23%
F
50%
66%
81%
Together our rate and equilibrium measurements on this
system demonstrate two important features of this reaction: (1)
the N-phenylmorpholine product arene binds with roughly
a Yield by GC-FID. b 14 mol% phosphine used.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
triflate-free 4-PF6 complex. In theory, protonation of the
aniline product could decrease product inhibition, though any
potential improvement would be counterbalanced by compet-
ing protonation of the more-basic morpholine nucleophile. In
practice, addition of triflic acid with or without added triethyl-
amine leads to reductions in yield (Table 5). Small amounts of
triflate ion appear to be beneficial22 (entries 1 vs. 5 and 5 vs.
6), though larger quantities of lithium triflate led to poorer
results. Thus it would appear that alternative approaches are
still necessary to address product inhibition if higher TONs
are desired.
Arene binding in 2 vs. 3. Owing to our inability to isolate the
fluorobenzene analogue of 3, the reactivity of this putative
intermediate can only be inferred by comparison to complex 1.
Fig. 5 ORTEP diagram of complex 4-OTf. Ellipsoids are shown at 50% We undertook a computational comparison of arene binding
probability.
thermodynamics using our experimentally-determined ener-
gies for the complex 1/2 pair as a benchmark. DFT calculations
(M06L/def2-SVP/TZVP) indicate that N-phenylmorpholine
binding by 1 is exergonic by −8.3 kcal mol−1 at 100 °C, which
Both the bis(phosphine) monohydride ligand set in 3 and
the κ3-phosphine in 2 provide monoanionic 5-electron donor
environments, a motif that is conserved in pentamethyl-
is in good agreement with our experimentally determined
value of −5.3 kcal mol−1 derived from the equilibrium con-
cyclopentadienyl and cyclopentadienyl catalysts reported by
stant at 100 °C. N-Phenylmorpholine displacement of fluoro-
benzene in the ruthenium hydride version of the catalyst to
Grushin12 and Williams16 respectively. Among published
give 3 is computed to be exergonic by −9.6 kcal mol−1. Thus
systems for π-arene catalyzed SNAr, only a recent report from
Shi diverges from this pattern by employing a dicationic ruthe-
the small difference in affinity for the N-phenylmorpholine
nium bis(phosphine) complex.13 This observation inspired the
and fluorobenzene arene pair, computed for 2 and 3, (ΔΔGcalc
= 1.3 kcal mol−1) predicts that the decyclometalated and cyclo-
preparation of complex 4-OTf (eqn (5) and Fig. 5), which con-
serves the hydrido bis-phosphino motif found in 3. Like 3, 4-
OTf catalyzes the amination of fluorobenzene by morpholine
in good, albeit not quantitative yield in the absence of addi-
metalated forms of the catalyst are subject to comparably
strong product arene binding.
tives (Table 5, entry 1). 4-OTf does outperform in situ-generated
conditions for PPh3 (Table 5 entry 1 versus Table 4). Thus 4-
OTf offers a convenient, single-component precatalyst that can
Conclusions
be prepared in a single step from a commercially-available
ruthenium source.
Role of acid additives. Having identified that 4-OTf is an
accessible, single-component catalyst with some room for
improvement versus 3, we examined the role of added
Brønsted acid and/or metal triflates with both 4-OTf and the
In summary, our examination of the Ru-catalyzed SNAr of
fluoroarenes has revealed new details that shed light on a very
rare example of catalytic nucleophilic aromatic substitution at
a π-arene. We have demonstrated an additive-dependent
switch in the identity of the resting state of the catalyst result-
ing from the ligand’s ability to bind in either a κ2 or cyclometa-
lated κ3 forms. Isolation of both catalytic intermediates in the
cyclometalated form has allowed us to estimate the difference
Table 5 Effect of acid and base additives on reaction yield
in the free energy of product N-phenylmorpholine binding vs.
fluorobenzene binding – a key consideration in the arene
exchange step necessary for catalytic turnover. These experi-
mental results are contextualized with DFT calculations
showing comparable binding affinities for the κ2 form of the
catalyst observed in the presence of silane additives.
Experimental measurements and predictions of binding
enthalpies quantify the severity of product inhibition encoun-
tered in this example of π-arene SNAr amination. We show that
ligand cyclometalation is not a determining factor in the
ability of this class of cationic ruthenium complexes to serve
as catalysts for SNAr. The silane additive previously hypoth-
esized to function to sequester fluoride ion appears to contrib-
ute to productive catalysis primarily through its ability to
modulate ligand cyclometalation, an observation which has
Entry
Cat.
Additives
% yielda
1
2
3
4
5
6
7
8
4-OTf
4-OTf
4-OTf
4-OTf
4-PF6
4-PF6
4-PF6
4-PF6
None
1 equiv. TfOH
72
3
1 equiv. TfOH, 1 equiv. Et3N
0.1 equiv. TfOH, 1 equiv. Et3N
None
0.1 equiv. LiOTf
0.2 equiv. LiOTf
1.0 equiv. LiOTf
41
71
51
60
53
30
a Yield by GC-FID.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
allowed us to employ a simple, single-component precatalyst 10 E. P. Kündig and A. Böttcher, Transition metal arene
for fluorobenzene amination. Attempts are currently underway
to translate these findings into the design of more-robust and
efficient catalysts for π-arene catalyzed SNAr.
π-complexes in organic synthesis and catalysis, Springer-
Verlag, Berlin; New York, 2004, p. 232.
11 R. P. Houghton, M. Voyle and R. Price, J. Chem. Soc., Perkin
Trans. 1, 1984, 925–931.
12 A. I. Konovalov, E. O. Gorbacheva, F. M. Miloserdov and
V. V. Grushin, Chem. Commun., 2015, 51, 13527–13530.
13 Q.-K. Kang, Y. Lin, Y. Li and H. Shi, J. Am. Chem. Soc., 2020,
142, 3706–3711.
Conflicts of interest
There are no conflicts to declare.
14 M. Otsuka, H. Yokoyama, K. Endo and T. Shibata, Synlett,
2010, 2601–2606.
15 M. Otsuka, K. Endo and T. Shibata, Chem. Commun., 2010,
46, 336–338.
16 J. W. Walton and J. M. J. Williams, Chem. Commun., 2015,
51, 2786–2789.
17 M. F. Semmelhack, A. Chlenov and D. M. Ho, J. Am. Chem.
Soc., 2005, 127, 7759–7773.
18 E. L. Muetterties, J. R. Bleeke and A. C. Sievert,
J. Organomet. Chem., 1979, 178, 197–216.
19 J. Takaya and J. F. Hartwig, J. Am. Chem. Soc., 2005, 127,
5756–5757.
Acknowledgements
The authors thank Don Stec and Markus Voehler for assistance
with NMR experiments. Funding from Vanderbilt University is
gratefully acknowledged. This material is based upon work
supported by the National Science Foundation under grant no.
CHE-1847813. Acknowledgment is made to the Donors of the
American Chemical Society Petroleum Research Fund for
partial support of this research.
20 W. Tian, A new process for the synthesis of morpholinyl-
benzenes, WO2001087865A1, 2001.
Notes and references
1 L. A. P. Kane-Maguire, E. D. Honig and D. A. Sweigart, 21 G. R. Brown, A. J. Foubister and P. D. Ratcliffe, Tetrahedron
Chem. Rev., 1984, 84, 525–543. Lett., 1999, 40, 1219–1222.
2 M. Rosillo, G. Domínguez and J. Pérez-Castells, Chem. Soc. 22 A. C. Sievert and E. L. Muetterties, Inorg. Chem., 1981, 20,
Rev., 2007, 36, 1589–1604. 489–501.
3 F. C. Pigge, R. Dhanya and D. C. Swenson, Organometallics, 23 T. G. Traylor, K. Stewart and M. Goldberg, J. Am. Chem.
2009, 28, 3869–3875. Soc., 1984, 106, 4445–4454.
4 M. E. O’Reilly, S. I. Johnson, R. J. Nielsen, W. A. Goddard 24 T. G. Traylor and K. Stewart, Organometallics, 1984, 3, 325–
and T. B. Gunnoe, Organometallics, 2016, 35, 2053–2056. 327.
5 R. M. Moriarty, U. S. Gill and Y. Y. Ku, J. Organomet. Chem., 25 M. L. H. Green, D. S. Joyner and J. M. Wallis, J. Chem. Soc.,
1988, 350, 157–190.
Dalton Trans., 1987, 2823–2830.
6 D. Astruc, Tetrahedron, 1983, 39, 4027–4095.
7 A. J. Pearson, P. Y. Zhu, W. J. Youngs, J. D. Bradshaw and
26 S. Zhang, J. K. Shen, F. Basolo, T. D. Ju, R. F. Lang, G. Kiss
and C. D. Hoff, Organometallics, 1994, 13, 3692–3702.
D. B. McConville, J. Am. Chem. Soc., 1993, 115, 10376– 27 J. W. Walton and L. A. Wilkinson, π-Coordinated arene
10377.
metal complexes and catalysis, in Organometallic Chemistry:
Volume 42, The Royal Society of Chemistry, 2019, vol. 42,
pp. 125–171.
8 R. M. Moriarty and U. S. Gill, Organometallics, 1986, 5, 253–
256.
9 F. Hossner and M. Voyle, J. Organomet. Chem., 1988, 347, 28 M. Utsunomiya and J. F. Hartwig, J. Am. Chem. Soc., 2004,
365–371.
126, 2702–2703.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020